In the flow of genetic information from the genome to the transcriptome, epigenetic regulation plays a critical role in modulating the expression of genotypes to phenotypes in a tissue specific and temporarily specific manner. Epigenetic regulation is a fundamental mechanism that involves not only in normal developmental processes but also in many human diseases. Epigenetic regulation can take place in the methylation of DNA, covalent modifications of histone, or interactions between nucleosome and DNA. DNA methylation is the most stable form of epigenetic modification that leads to transcriptional silencing, X chromosome inactivation and imprinting. Obtaining genome-scale patterns of DNA methylation in different normal and disease tissues are critical for understanding the developmental processes as well as the etiology of many human diseases However, in contrast to rapid advances in human genome sequencing, it is still impractical to characterize the methylation status of every single CpG in the human genome at a reasonable cost. The goal of this proposal is to enable digital quantification of DNA methylation status of non-repetitive CpG islands on the genome scale efficiently and inexpensively. We plan to achieve this goal by developing two complementary methods that can specifically extract all non-repetitive CpG islands (or any subset) from the bisulfite converted genome, and coupling these target selection methods with next-generation DNA sequencing technologies. This will lead to a significant reduction (~100-fold) in the cost of sequencing, as the size of the sequencing target is reduced from the full human genome to all non-repetitive CpG islands of ~20Mbps in total length. In addition, since all CpG islands will be captured and sequenced as a pool through a series of single-tube reactions, our methods do not require robotic devices for complicated liquid handling. The simplicity and scalability of our methods will enable true genome-scale analysis of DNA methylation patterns in a large number of biological samples. Recognizing the rapid evolution of DNA sequencing technology, we will also optimize the target selection front-ends to be widely adaptable to different DNA sequencing platforms, such that the throughput and cost will scale linearly in the future. The specific aims are as follows: 1) Single molecule bisulfite sequencing of all non-repetitive CpG islands in the human genome using padlock probes; 2) array-based capture of CpG islands for shotgun bisulfite sequencing of all non-repetitive CpG islands in the human genome; 3)characterization of global DNA methylation changes in the differentiation of stem cells; 4) develop computational infractructure and methods that will enable users to easily acquire, evaluate and visualize the next-generation DNA sequencing data and will help optimize the bisulfite sequencing technology development. The proposed genome-scale bisulfite sequencing approach will represent significant improvements over existing epigenomic profiling technologies in both the scale (>80% CpG islands in the genome) and the resolution (single CpG, single molecule) of analyses. It will accelerate the studies of global DNA methylation patterns in various tissues at different developmental stages. It will also provide digital profiles of aberrant DNA methylation in many human diseases and offer a robust method for classifying disease subtypes. [unreadable] [unreadable] PUBLIC HEALTH RELEVANCE: Epigenetic processes modulate the packaging and function of the human genome in normal developmental processes and many pathologic states, including human cancers. We propose to develop genome-scale bisulfite genomic sequencing methods for global digital analysis of DNA methylation, as well as the associated computational methods for the analysis and visualization of the massive bisulfite sequencing data. Through the seamless integration of targeted epigenomic capture and next-generation DNA sequencing, we will enable genome-scale digital profiling of the DNA methylation landscape across the human genome. This enabling technology will help understand the functional roles of DNA methylation in gene regulation in various developmental processes and human diseases. [unreadable] [unreadable] [unreadable]